Channel Modeling and Signal Processing for Underwater Massive MIMO System Under Link Misalignment

Aiming at the problem of increased interference of different light signals caused by the increase in the number of LEDs, the link misalignment of the underwater optical wireless communication (UOWC) system, and the shift and expansion of the light spot in the massive multiple in multiple out (Massive-MIMO) optical communication system, this paper uses an aspherical lens set as the receiving optical antenna to separate different beams and optimizes the lens structure to further completely separate different beams. In this paper, the Singular value decomposition based on the least-squares channel estimation (LS-SVD) precoding algorithm is used to reduce the impact of communication link misalignment on the performance of the communication system and improve the system bit error rate performance (BER). The simulation results show that the LS-SVD precoding algorithm is better than the traditional Lattice Reduction (LR) precoding algorithm, zero-forcing (ZF) detection algorithm, and the minimum mean square error (MMSE) detection algorithm under the condition of misalignment of the communication link in the underwater Massive-MIMO system, the LS-SVD precoding algorithm improves the BER by about an order of magnitude under the same signal-to-noise ratio (SNR).